DC Branches

DC Branches represent overhead lines and cables in DC connections and DC grids.

Parameters

Set of parameters used to model DC branches as defined in the input data

namesymb.unittypedefaultdefinition
index$d$-Int-unique index of the DC branch
f_bus$i$-Int-unique index of the bus to which the DC branch is originating from
t_bus$j$-Int-unique index of the bus to which the DC branch is terminating at
r$r_{dc}$p.u.Real-resistance of the DC branch
l$l_{dc}$p.u.Real-inductance of the DC branch (not used)
c$c_{dc}$p.u.Real-capacitance of the DC branch (not used)
rateA$\overline{P_{dc}}$MWReal-long term rating of the DC branch
rateB$\overline{P^{st}_{dc}}$MWReal-short term term rating of the DC branch
rateC$\overline{P^{em}_{dc}}$MWReal-emergency rating of the DC branch
status$\delta_{dc}$-Int-status indicator of the DC branch
p$n_{p}$-Int-number of DC poles, specified at test case level
cost$C_{dc}$-Real-investment cost for DC branch used in TNEP problems

Variables

Optimisation variables representing DC branch behaviour

namesymb.unitformulationdefinition
p_dcgrid$P_{d,i,j}$p.u.ACP, ACR, LPAC, IVR, SOC, DCP, NFActive power flow through DC branch d, connecting DC nodes i and j
igrid_dc$I_{d,i,j}$p.u.IVRCurrent flow through DC branch d, connecting DC nodes i and j
ccm_dcgrid$J_{d,i,j}$p.u.SOC, QCSquare of current flow through DC branch d, connecting DC nodes i and j
wdcr$W_{d,i,j}$p.u.SOC, QCLifted variable representing bilinear voltage product ($V_{i} \cdot V_{j}$) of branch d, connecting DC nodes i and j

Constraints

Flow, current and voltage product limits

\[\begin{align} - \overline{P_{dc}} &\leq P_{d,i,j} \leq \overline{P_{dc}} \\ - \overline{I_{dc}} &\leq I_{d,i,j} \leq \overline{I_{dc}} \\ 0 &\leq J_{d,i,j} \leq \overline{I_{dc}^{2}} \\ 0 &\leq W_{d,i,j} \leq max(V_{i}^{2}, V_{j}^{2}) \end{align}\]

DC branch admittance

\[g_{dc} = \frac{1}{r_{dc}}\]

Ohm's law

ACP, ACR model:

\[\begin{align} P_{d,i,j} &= n_{p} \cdot g_{dc} \cdot V_{i} \cdot (V_{i} - V_{j}) \\ P_{d,j,i} &= n_{p} \cdot g_{dc} \cdot V_{j} \cdot (V_{j} - V_{i}) \end{align}\]

SOC, QC bus injection model (BIM):

\[\begin{align} P_{d,i,j} &= n_{p} \cdot g_{dc} \cdot (W_{i} - W_{d,i,j}) \\ P_{d,j,i} &= n_{p} \cdot g_{dc} \cdot (W_{j} - W_{d,i,j}) \end{align}\]

SOC, QC branch flow model (BFM):

\[\begin{align} W_{j} &= W_{i} - \frac{2 \cdot r_{dc} \cdot P_{d,i,j}} {n_{p}} + r_{dc}^{2} \cdot J_{d,i,j} \\ P_{d,i,j} + P_{d,j,i} &= r_{dc} \cdot n_{p} \cdot J_{d,i,j} \\ P_{d,i,j}^{2} &\leq n_{p}^{2} \cdot W_{i} \cdot J_{d,i,j} \end{align}\]

IVR model:

\[\begin{align} V_{j} = V_{i} - \frac{r_{dc} \cdot I_{d,i,j}}{n_{p}} \\ V_{i} = V_{j} - \frac{r_{dc} \cdot I_{d,j,i}}{n_{p}} \\ P_{d,i,j} = I_{d,i,j} \cdot V_{i} \\ P_{d,j,i} = I_{d,j,i} \cdot V_{j} \end{align}\]

DCP and NF model:

In this model there are no losses, as such only the active power limits are binding.